Comment on Misra et al. Homozygous Hypomorphic HNF1A Alleles Are a Novel Cause of Young-Onset Diabetes and Result in Sulfonylurea-Sensitive Diabetes. Diabetes Care 2020;43:909–912

With great interest, we have read the article by Misra et al. (1), which reported a family that suffered from young-onset and sulfonylurea-sensitive diabetes resulting from the homozygous hypomorphic HNF1A mutation. We appreciate the considerable achievement of new insights in diagnosis and pathogenicity of HNF1A germline mutation–induced diabetes, but we would like to make some comments about the interpretation of the results and discussion.

Misra et al. reported that the proband’s father met the criteria of prediabetes and the mother developed diabetes during pregnancy at the age of 29 years. The authors also concluded that two reported studies (2,3) featured heterozygous p.A251T HNF1A variant carriers with older onset of diabetes, perhaps higher BMI, and milder symptoms than the proband with the homozygous mutation. Moreover, the heterozygous father had a high-sensitivity C-reactive protein (hs-CRP) level of 1.4 mg/L, which might indicate low-grade systemic inflammation and could be explained by his age and BMI from the authors’ view. However, low circulating hs-CRP levels are associated with loss of function or defective function of HNF1A variants because the CRP gene has HNF1A binding sites on its promoter (4). We also observed HNF1A-MODY (maturity-onset diabetes of the young) families and found low level of hs-CRP with a heterozygous p.T260M HNF1A mutation carrier with depression (5). Given the clinical manifestation of the family, including higher BMI, C-peptide, and hs-CRP levels, we could not exclude that the situation resulted from the cumulative effect of polygenic mutations in metabolic derangement or diabetes instead of monogenic mutation of the heterozygous p.A251T HNF1A in these studies. In particular, Misra et al. and researchers of two studies that reported heterozygous p.A251T HNF1A variant only sequenced the HNF1A gene, with or without testing 22 monogenic diabetes candidate genes, instead of conducting whole genome or exome sequencing. Another piece of supporting evidence was that authors marked II-4’s (proband’s sister 2) indexed box black on the family pedigree, which indicated that II-4 also met the criteria of diabetes and her age at diagnosis was 18 years. She has a higher chance of carrying the homozygous mutation; it is unclear how they could explain the impossible events on a frequency of 0.004 (0.25⁴) or 0.016 (0.25³) when obeying the law of Mendelian inheritance of monogenic diabetes.

Of interest to us was the BMI among the three homozygous and six heterozygous p.A251T HNF1A mutation participants. There is an obvious tendency of lower BMI at onset compared with early-onset familial type 2 diabetes in Chinese HNF1A-MODY patients from our study. We suggest that BMI before diabetes diagnosis or at onset of diabetes is more representative in a metabolic state of monogenic diabetes than BMI after patients have been diagnosed and treated for diabetes for a period. Glycometabolism derangement, complications, antidiabetic agents, and lifestyle intervention could affect BMI. More data about BMI before diabetes diagnosis or at onset of diabetes could help us to make an initial differential diagnosis before genetic testing.